Skip to content
Submit incident
ActiveAdopted and in force, or in effect for the bodies it covers.

Executive Order on Promoting the Use of Trustworthy Artificial Intelligence in the Federal Government

United States Β· Regulation Β· The President of the United States

Our summary

This executive order directs federal agencies to adopt and use artificial intelligence responsibly while fostering public trust and protecting civil rights and liberties. It establishes nine principles for AI use in government, including requirements for lawfulness, accuracy, security, transparency, and accountability. The order requires agencies to inventory their AI use cases, comply with the principles, and coordinate implementation across government.

Written by the Index in our own words. It is not legal advice; the official text is the authority.

Inside this policy

Every section of this policy in order. Our notes summarise each one in plain language and do not quote the official text, which is the authority.

Drafted by the Index An automated summary of the official text, not yet reviewed by a person.Reviewed Checked by a person against the official text.

10 sections

  1. Section 1 Purpose

    Federal agencies are encouraged to adopt AI to improve operations and service delivery while ensuring such applications protect personal information, individual rights and freedoms, and foundational national principles. The executive order directs agencies to implement guiding principles and governance mechanisms for non-defense AI uses to maintain public confidence in government systems and alignment with American values.

    Drafted by the Index
  2. Section 2 Policy

    Declares that the United States will promote AI adoption by responsible agencies to enhance government operations while fostering public trust and protecting civil liberties, guided by principles set forth in the order and governed by OMB policy guidance.

    Drafted by the Index
  3. Section 3 Principles for Use of AI in Government

    Requires agencies to ensure their AI systems are lawful and respectful of national values; purposeful with benefits outweighing risks; accurate and reliable for their intended use cases; safe and resilient against threats; understandable to relevant experts and users; managed with clear human responsibility; regularly tested and monitored; transparent to appropriate stakeholders; and accountable through proper safeguards and personnel training.

    Drafted by the Index
  4. Section 4 Implementation of Principles

    Directs OMB to issue a public roadmap within 180 days for creating or revising policy guidance on AI use in government, considering voluntary consensus standards, while existing relevant OMB policies continue to apply to AI use.

    Drafted by the Index
  5. Section 5 Agency Inventory of AI Use Cases

    Requires the federal CIO Council to establish criteria for agency AI inventories within 60 days; each agency must then compile its unclassified AI uses within 180 days and thereafter annually, identify inconsistencies with the order, develop correction or retirement plans within 120 days, share inventories with other agencies, and make them publicly available while protecting sensitive information.

    Drafted by the Index
  6. Section 6 Interagency Coordination

    Expects agencies to participate in interagency bodies to advance implementation of AI principles, with the CIO Council to publish a list of recommended forums within 45 days.

    Drafted by the Index
  7. Section 7 AI Implementation Expertise

    Directs the Presidential Innovation Fellows program to establish an AI track within 90 days to recruit industry and academic experts to federal agencies; requires OPM to inventory federal rotational programs within 45 days and issue recommendations within 180 days on using these programs to build AI expertise in agencies.

    Drafted by the Index
  8. Section 8 Responsible Agencies and Officials

    Defines the order as applying to federal agencies with AI use cases, excluding the Department of Defense and Intelligence Community agencies; requires each agency to designate responsible officials within 30 days to coordinate implementation of the principles.

    Drafted by the Index
  9. Section 9 Scope of Application

    Defines which artificial intelligence systems are covered by this order: AI used to carry out agency missions, enhance decision making, or deliver public benefits, including both standalone and embedded systems developed internally or by third parties. Excludes AI in defense and national security systems, AI components within commercial products like word processors or navigation tools, and AI research activities, though agencies must still follow other applicable guidelines and comply with law when using commercial products.

    Drafted by the Index
  10. Section 10 General Provisions

    Clarifies that the order does not diminish the legal authority of executive agencies or the Director's budget and legislative functions, must be carried out according to law and available funding, and does not create enforceable rights for any party.

    Drafted by the Index

Controls

What this policy requires, forbids, grants or penalises, one control at a time and in our own words. Each control links to the official text, which is the authority. This is not legal advice.

Drafted by the Index Drafted automatically from the official text, not yet reviewed by a person.Reviewed Checked by a person against the official text.

Themes covered 12 of 35

Choose a theme to see only its controls. About the themes

Risk and scope 2 of 4

  • Prohibited practices, not covered
  • Impact assessment, not covered

Data and privacy 1 of 4

  • Data governance and quality, not covered
  • Cross-border data transfer, not covered
  • Copyright and training data, not covered

Transparency and records 2 of 4

  • Labelling of synthetic content, not covered
  • Technical documentation, not covered

Technical safeguards 3 of 4

  • Bias and non-discrimination, not covered

Assurance and lifecycle 1 of 4

  • Conformity assessment and certification, not covered
  • Registration and notification, not covered
  • Incident and breach reporting, not covered

Rights and protected groups 0 of 4

  • Automated decision-making rights, not covered
  • Children and vulnerable groups, not covered
  • Biometric identification and surveillance, not covered
  • Whistleblowing, complaints and redress, not covered

Roles and organisation 3 of 6

  • Supply chain and provider-deployer duties, not covered
  • General purpose and foundation models, not covered
  • Innovation support and sandboxes, not covered

Enforcement and sectors 0 of 5

  • Enforcement, penalties and regulators, not covered
  • Sector rules: health, not covered
  • Sector rules: finance and insurance, not covered
  • Sector rules: employment, not covered
  • Sector rules: education, not covered

36 controls

Group by

Section 2 3 controls

  1. ObligationApplies to: Public authorityDrafted by the Index

    The United States will encourage responsible agencies to build and deploy AI systems that enhance federal operations while safeguarding public confidence, national interests, and compliance with applicable privacy, civil rights and civil liberties statutes.

    Section 2
  2. ObligationApplies to: Public authorityDrafted by the Index

    Responsible agencies shall apply the principles articulated in section 3 when creating, procuring or operating AI systems.

    Section 2
  3. ObligationApplies to: Public authorityDrafted by the Index

    The Office of Management and Budget shall issue binding policy guidance to govern how agencies employ AI principles.

    Section 2

Section 3 8 controls

  1. ObligationApplies to: Public authorityDrafted by the Index

    Agencies must build and deploy AI systems in ways that respect the Constitution and comply with all applicable law governing privacy, civil rights and civil liberties.

    Section 3
  2. ObligationApplies to: Public authorityDrafted by the Index

    Agencies shall pursue AI applications only when the advantages clearly exceed the risks and those risks can be identified and mitigated.

    Section 3
  3. ObligationApplies to: Public authorityDrafted by the Index

    Agencies must verify that AI systems function accurately, reliably and as intended for their specified operational contexts.

    Section 3
  4. ObligationApplies to: Public authorityDrafted by the Index

    Agencies shall protect their AI systems from systematic weaknesses, hostile tampering and other harmful threats.

    Section 3
  5. ObligationApplies to: Public authorityDrafted by the Index

    Agencies shall design AI systems so that specialists, operators and other relevant persons can understand how they function and what they produce.

    Section 3
  6. ObligationApplies to: Public authorityDrafted by the Index

    Agencies must define and assign human roles and responsibilities for all phases of AI creation and use, and document the design, inputs and outputs of each system.

    Section 3
  7. ObligationApplies to: Public authorityDrafted by the Index

    Agencies shall repeatedly evaluate their AI systems against these principles and maintain mechanisms to stop or retire systems that deviate from their intended purpose.

    Section 3
  8. ObligationApplies to: Public authorityDrafted by the Index

    Agencies shall tell Congress and the public about their use of AI as far as laws and policies permit.

    Section 3

Section 4 3 controls

  1. ObligationApplies to: Public authorityDrafted by the Index

    Existing OMB policies that address information technology design, creation, procurement and use shall continue to govern those aspects of AI use unless they conflict with this order or applicable law.

    Section 4
  2. ObligationApplies to: Public authorityDrafted by the Index

    The OMB Director shall release a public roadmap within 180 days outlining what new or revised policy guidance on federal AI use will be created and when.

    Section 4
  3. ObligationApplies to: Public authorityDrafted by the Index

    Agencies shall employ voluntary industry standards for AI systems and processes where they align with law and practical operations, and OMB shall consider such standards when drafting future guidance.

    Section 4

Section 5 6 controls

  1. ObligationApplies to: Public authorityDrafted by the Index

    The Federal CIO Council shall establish and publicize requirements and procedures for documenting agency AI uses within 60 days.

    Section 5
  2. ObligationApplies to: Public authorityDrafted by the Index

    Each agency shall compile and maintain an inventory of its unclassified and non-sensitive AI applications within 180 days, then update it annually.

    Section 5
  3. ObligationApplies to: Public authorityDrafted by the Index

    Agencies shall examine their existing AI systems to detect whether they breach this order's requirements.

    Section 5
  4. ObligationApplies to: Public authorityDrafted by the Index

    For AI applications that do not comply, agencies must create a plan within 120 days to bring them into line with the order or remove them, and the agency's responsible official must approve the plan within that period.

    Section 5
  5. ObligationApplies to: Public authorityDrafted by the Index

    Agencies shall work with the Agency Data Governance Body and other relevant offices to put their correction or retirement plans into effect within 180 days of approval, subject to available resources.

    Section 5
  6. ObligationApplies to: Public authorityDrafted by the Index

    Agencies shall publish their inventories to the public within 120 days of completing them, except for data that is private, sensitive or restricted under applicable statutes and directives.

    Section 5

Section 6 2 controls

  1. ObligationApplies to: Public authorityDrafted by the Index

    Federal agencies should take part in cross-agency bodies to help fulfill the Principles and promote trustworthy AI in line with this order.

    Section 6
  2. ObligationApplies to: Public authorityDrafted by the Index

    The CIO Council must publish a curated list of interagency forums that agencies may join based on their authorities and missions.

    Section 6

Section 7 3 controls

  1. ObligationApplies to: Public authorityDrafted by the Index

    Within 90 days of the order, the Presidential Innovation Fellows program must identify priority expertise areas and establish an AI track to bring industry and academic experts to federal agencies.

    Section 7
  2. ObligationApplies to: Public authorityDrafted by the Index

    Within 45 days of the order, the Office of Personnel Management must create an inventory of federal rotational programs and assess how they can expand AI expertise in agencies.

    Section 7
  3. ObligationApplies to: Public authorityDrafted by the Index

    Within 180 days of creating the rotational programs inventory, OPM must issue recommendations on how to use these programs to expand AI expertise in agencies and share findings with interagency bodies.

    Section 7

Section 8 1 control

  1. ObligationApplies to: Public authorityDrafted by the Index

    Within 30 days of the order, each covered agency must designate responsible officials to coordinate implementation of the order's principles with the Agency Data Governance Body and interagency coordination bodies.

    Section 8

Section 9 6 controls

  1. ObligationApplies to: Public authorityDrafted by the Index

    The Principles and implementation guidance apply to AI that agencies design, develop, buy or use to execute their missions, improve decision-making, or benefit the public.

    Section 9
  2. ObligationApplies to: Public authorityDrafted by the Index

    The order covers both new and existing AI systems, whether standalone or built into other tools, developed in-house or by contractors, along with their training data and outputs.

    Section 9
  3. ObligationApplies to: Public authorityDrafted by the Index

    The order applies to agency purchases of AI tools and applications.

    Section 9
  4. PermissionApplies to: Public authorityDrafted by the Index

    The order does not apply to AI systems used for defense or national security, but agencies must follow other relevant rules and standards in those areas.

    Applies when: defense or national security systems

    Section 9
  5. PermissionApplies to: Public authorityDrafted by the Index

    The order exempts AI embedded in ordinary business software such as text editors or navigation tools, though government use of such products must still meet legal and policy requirements on safety, security, privacy and civil rights.

    Applies when: AI embedded within commercial products

    Section 9
  6. PermissionApplies to: Public authorityDrafted by the Index

    AI research and development work is excluded from the order, though the Principles should shape future federal AI research.

    Applies when: research and development activities

    Section 9

Section 10 4 controls

  1. PermissionApplies to: Public authorityDrafted by the Index

    This order does not diminish the statutory powers of federal departments, agencies or their heads.

    Section 10
  2. PermissionApplies to: Public authorityDrafted by the Index

    This order does not limit the Director of OMB's budget-related or legislative functions.

    Section 10
  3. ObligationApplies to: Public authorityDrafted by the Index

    Implementation of this order must follow all applicable laws and depend on congressional appropriations.

    Section 10
  4. ProhibitionApplies to: OthersDrafted by the Index

    This order creates no enforceable legal rights, remedies or standing for any individual or organisation against the United States or its officials.

    Section 10

Change history

No changes recorded since we began monitoring this policy.